FeDABoost: Fairness Aware Federated Learning with Adaptive Boosting
This addresses fairness and performance issues in Federated Learning for non-IID data, but it is incremental as it builds on existing methods like FedAvg and Ditto with adaptive modifications.
This paper tackled the problem of improving performance and fairness in Federated Learning under non-IID settings by proposing FeDABoost, which integrates dynamic boosting and adaptive gradient aggregation. The results show that FeDABoost achieves improved fairness and competitive performance on benchmark datasets like MNIST, FEMNIST, and CIFAR10 compared to FedAvg and Ditto.
This work focuses on improving the performance and fairness of Federated Learning (FL) in non IID settings by enhancing model aggregation and boosting the training of underperforming clients. We propose FeDABoost, a novel FL framework that integrates a dynamic boosting mechanism and an adaptive gradient aggregation strategy. Inspired by the weighting mechanism of the Multiclass AdaBoost (SAMME) algorithm, our aggregation method assigns higher weights to clients with lower local error rates, thereby promoting more reliable contributions to the global model. In parallel, FeDABoost dynamically boosts underperforming clients by adjusting the focal loss focusing parameter, emphasizing hard to classify examples during local training. We have evaluated FeDABoost on three benchmark datasets MNIST, FEMNIST, and CIFAR10, and compared its performance with those of FedAvg and Ditto. The results show that FeDABoost achieves improved fairness and competitive performance.